Hi Dave;
Answers below.
On Thu, Feb 3, 2011 at 10:48 PM, Dave <[email protected]> wrote:
> If you are just counting the number of rotations per minute or second I
> don't think you are going to get what you want.
>
> You really need to be able to detect a phase change so you need to know
> the generator rotor position - at least a lot more accurate than one
> phase rotation at 60hz.
>
> The error you are seeing is 1/60 of a second which is the minimum error
> you can see. Your current system will probably hunt between an error
> of zero and 1.
>
> You need finer resolution.
>
> A second is really quite a long time in this situation.
>
> A hall effect device watching holes fly by on a pulley should work or
> watching gear teeth. A coarse encoder would also work.
>
I'm convinced. I plan to use approximately 10000 teeth per second.
>
> What are you using for your reference timebase as the speed setpoint for
> the generator?? How accurate do you want this generator to run?
>
Plan to use the computer clock synchronised by one of the server time
correction apps for Linux.
>
> Do you want to be able to switch it in on out from a grid connection?
>
No, it is for off grid. With the computer clock correct all the timers and
AC clocks should average to the correct time.
Thank you very much for your help
Don
>
> Dave
>
> On 2/3/2011 9:32 PM, Jon Elson wrote:
> > Don Stanley wrote:
> >
> >> Hi All;
> >> I am attempting to configure EMC2 10.04.06 to control RPM
> >> on a 60 HZ AC generator.
> >>
> >> The basic approach is:
> >> -Set the steps per inch to the number of AC cycles per day.
> >> -Monitor the AC frequency into phase A of a HAL encoder.
> >> -Feed the encoder output into the axis and PID feedback.
> >> -Feed the axis motor-pos-cmd to stepgen position-cmd.
> >> -Feed the stepgen out through the paraport to a stepper motor driver.
> >>
> >> This seems to work in general but there are two problems I
> >> have not been able to pin down.
> >> 1- The PID error is a sawtooth shape causing the throttle
> >> stepper motor to quiver at 60 HZ.
> >> A HAL Scope image is available at
> >> http://imagebin.org/135986
> >>
> >>
> >>
> > The sawtooth appears to be happening at 16.6 ms, which is one 60 Hz
> cycle.
> > Perhaps that is significant. What is the velocity coming out of the
> > encoder counter
> > component? Does it waver at all? If all these calculations are
> > performed in arithmetic
> > scaled to days, you may be looking at the limits of binary precision. I
> > notice your
> > sawtooth is fluctuating by much less than 1 millionth of a unit, roughly
> > 2 x 10 ^ -7.
> >
> > Jon
> >
> >
> ------------------------------------------------------------------------------
> > The modern datacenter depends on network connectivity to access resources
> > and provide services. The best practices for maximizing a physical
> server's
> > connectivity to a physical network are well understood - see how these
> > rules translate into the virtual world?
> > http://p.sf.net/sfu/oracle-sfdevnlfb
> > _______________________________________________
> > Emc-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> >
>
>
>
> ------------------------------------------------------------------------------
> The modern datacenter depends on network connectivity to access resources
> and provide services. The best practices for maximizing a physical server's
> connectivity to a physical network are well understood - see how these
> rules translate into the virtual world?
> http://p.sf.net/sfu/oracle-sfdevnlfb
> _______________________________________________
> Emc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
------------------------------------------------------------------------------
The modern datacenter depends on network connectivity to access resources
and provide services. The best practices for maximizing a physical server's
connectivity to a physical network are well understood - see how these
rules translate into the virtual world?
http://p.sf.net/sfu/oracle-sfdevnlfb
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
